Secondary organic aerosol (SOA), a major atmospheric pollutant impacting climate and human health, is significantly underestimated by air quality models, due to the poorly understood formation and evolution of SOA in the atmosphere, as well as the uncertainties in the emission of SOA precursors. The aforementioned limited understandings hinder the efforts to mitigate the adverse effects.

The aim of this dissertation is to evaluate the SOA formation from the oxidation of semi-/intermediate-volatile organic compounds (IVOC/SVOC), which are recently identified as major SOA contributors undergoing oxidation, dilution, gas-particle partition, and aging processes. The Volatility Basis Set (VBS) models incorporating these processes have been implemented into the CAMx (Comprehensive Air Quality Model with Extends), by mapping SOA precursors including primary organic aerosol (POA) into a series volatility bins with parameterized chemical and physical properties (e.g. molecular weight, oxidation state), and by incorporating the aging processes such as oxidations with OH radical.

Different VBS models are evaluated with CAMx simulation: 1-D and 1.5-D, of which results show that CAMx with 1-D VBS model provides temporal variations and concentration levels of SOA with generally better agreement with observations from the high-resolution time-of-flight aerosol mass spectrometry (HR-ToF-AMS) over urban, rural, and coastal sites in the Pearl River Delta (PRD) region, China, than other SOA treatment methods such as traditional SOA module and 1.5-D VBS model. Furthermore, one year simulation of 1-D VBS model is conducted and indicates that the oxidation of IVOC/SVOC contributes SOA concentration as high as 2-7 ug/m³ (50%-70% of total SOA concentration) in the PRD region on annual average. Based on better understanding about SOA formation supported by VBS models, SOA source apportionment technology (SSAT) is developed to track the dispersion of SOA, and the results in the PRD region suggest that regional transport and super-regional transport play the most important roles in building up SOA concentration.

Results in this dissertation contribute to enlightening our understanding of the physical and chemical processes, that SOA and their precursors are involved in atmospheric environments. Moreover, the applications of VBS models and SSAT algorithm provide new insights into and policy implications for improving local and regional air quality.